Resources / Topic 02
Open-source approaches to AI-assisted security audits, vulnerability discovery, remediation, and verification.
5 resources
Explore multi-stage agent workflows for finding, validating, reporting, and remediating code vulnerabilities.
A multi-phase coding-agent skill for recon, vulnerability hunting, adversarial validation, structured findings, and independent verification.
Best for Studying how to turn a coding agent into a disciplined security-audit workflow for authorized codebases.
OpenAI's Apache-2.0 CLI and TypeScript SDK that scans a codebase for vulnerabilities, compares findings across scans, and proposes fixes.
Best for Wiring model-driven scans into a DevSecOps loop and tracking regressions between runs; scan only repositories you are authorized to test, and review every finding and fix before acting on it.
An autonomous, white-box AI pentester for web applications and APIs that combines source analysis with live exploitation to validate exploitable vulnerabilities.
Best for Studying proof-by-exploitation workflows; run only against sandboxed systems you own or have explicit written authorization to test.
A Snyk scanner that inventories AI agents, MCP servers, and agent skills, checking for prompt injection and other agent supply-chain threats.
Best for Auditing local agent components before use; scan untrusted MCP configurations inside a sandbox and review commands before consent.
An open-source agentic SAST pipeline covering threat modeling, multi-lens analysis, adversarial verification, remediation, and fix validation.
Best for Understanding an end-to-end vulnerability workflow; scan only authorized repositories, and review every generated finding or fix before adoption.
Nothing matches this filter.